summaryrefslogtreecommitdiffstats
path: root/llvm/lib/Transforms/InstCombine
Commit message (Collapse)AuthorAgeFilesLines
* InstCombine: (icmp eq B, 0) | (icmp ult A, B) -> (icmp ule A, B-1)David Majnemer2013-07-051-1/+28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | This transform allows us to turn IR that looks like: %1 = icmp eq i64 %b, 0 %2 = icmp ult i64 %a, %b %3 = or i1 %1, %2 ret i1 %3 into: %0 = add i64 %b, -1 %1 = icmp uge i64 %0, %a ret i1 %1 which means we go from lowering: cmpq %rsi, %rdi setb %cl testq %rsi, %rsi sete %al orb %cl, %al ret to lowering: decq %rsi cmpq %rdi, %rsi setae %al ret llvm-svn: 185677
* InstCombine: Reimplementation of visitUDivOperandDavid Majnemer2013-07-041-56/+139
| | | | | | | | | | | This transform was originally added in r185257 but later removed in r185415. The original transform would create instructions speculatively and then discard them if the speculation was proved incorrect. This has been replaced with a scheme that splits the transform into two parts: preflight and fold. While we preflight, we build up fold actions that inform the folding stage on how to act. llvm-svn: 185667
* Use SmallVectorImpl::iterator/const_iterator instead of SmallVector to avoid ↵Craig Topper2013-07-041-1/+1
| | | | | | specifying the vector size. llvm-svn: 185606
* Revert r185257 (InstCombine: Be more agressive optimizing 'udiv' instrs with ↵Hal Finkel2013-07-021-77/+44
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 'select' denoms) I'm reverting this commit because: 1. As discussed during review, it needs to be rewritten (to avoid creating and then deleting instructions). 2. This is causing optimizer crashes. Specifically, I'm seeing things like this: While deleting: i1 % Use still stuck around after Def is destroyed: <badref> = select i1 <badref>, i32 0, i32 1 opt: /src/llvm-trunk/lib/IR/Value.cpp:79: virtual llvm::Value::~Value(): Assertion `use_empty() && "Uses remain when a value is destroyed!"' failed. I'd guess that these will go away once we're no longer creating/deleting instructions here, but just in case, I'm adding a regression test. Because the code is bring rewritten, I've just XFAIL'd the original regression test. Original commit message: InstCombine: Be more agressive optimizing 'udiv' instrs with 'select' denoms Real world code sometimes has the denominator of a 'udiv' be a 'select'. LLVM can handle such cases but only when the 'select' operands are symmetric in structure (both select operands are a constant power of two or a left shift, etc.). This falls apart if we are dealt a 'udiv' where the code is not symetric or if the select operands lead us to more select instructions. Instead, we should treat the LHS and each select operand as a distinct divide operation and try to optimize them independently. If we can to simplify each operation, then we can replace the 'udiv' with, say, a 'lshr' that has a new select with a bunch of new operands for the select. llvm-svn: 185415
* InstCombine: Also turn selects fed by an and into arithmetic when the types ↵Benjamin Kramer2013-06-291-4/+9
| | | | | | | | | don't match. Inserting a zext or trunc is sufficient. This pattern is somewhat common in LLVM's pointer mangling code. llvm-svn: 185270
* InstCombine: FoldGEPICmp shouldn't change sign of base pointer comparisonDavid Majnemer2013-06-291-2/+1
| | | | | | | | | | | | | | | | | | | | | | | | Changing the sign when comparing the base pointer would introduce all sorts of unexpected things like: %gep.i = getelementptr inbounds [1 x i8]* %a, i32 0, i32 0 %gep2.i = getelementptr inbounds [1 x i8]* %b, i32 0, i32 0 %cmp.i = icmp ult i8* %gep.i, %gep2.i %cmp.i1 = icmp ult [1 x i8]* %a, %b %cmp = icmp ne i1 %cmp.i, %cmp.i1 ret i1 %cmp into: %cmp.i = icmp slt [1 x i8]* %a, %b %cmp.i1 = icmp ult [1 x i8]* %a, %b %cmp = xor i1 %cmp.i, %cmp.i1 ret i1 %cmp By preserving the original sign, we now get: ret i1 false This fixes PR16483. llvm-svn: 185259
* InstCombine: Small whitespace cleanup in FoldGEPICmpDavid Majnemer2013-06-291-1/+1
| | | | llvm-svn: 185258
* InstCombine: Be more agressive optimizing 'udiv' instrs with 'select' denomsDavid Majnemer2013-06-291-44/+77
| | | | | | | | | | | | | | | | | Real world code sometimes has the denominator of a 'udiv' be a 'select'. LLVM can handle such cases but only when the 'select' operands are symmetric in structure (both select operands are a constant power of two or a left shift, etc.). This falls apart if we are dealt a 'udiv' where the code is not symetric or if the select operands lead us to more select instructions. Instead, we should treat the LHS and each select operand as a distinct divide operation and try to optimize them independently. If we can to simplify each operation, then we can replace the 'udiv' with, say, a 'lshr' that has a new select with a bunch of new operands for the select. llvm-svn: 185257
* InstCombine: Optimize (1 << X) Pred CstP2 to X Pred Log2(CstP2)David Majnemer2013-06-281-2/+72
| | | | | | | | | | | | | | We may, after other optimizations, find ourselves with IR that looks like: %shl = shl i32 1, %y %cmp = icmp ult i32 %shl, 32 Instead, we should just compare the shift count: %cmp = icmp ult i32 %y, 5 llvm-svn: 185242
* Fix using arg_end() - arg_begin() instead of arg_size()Matt Arsenault2013-06-281-3/+3
| | | | llvm-svn: 185121
* Revert "Revert "[APFloat] Removed APFloat constructor which initialized to ↵Michael Gottesman2013-06-271-4/+4
| | | | | | | | | | | | | | | | | | | | either zero/NaN but allowed you to arbitrarily set the category of the float."" This reverts commit r185099. Looks like both the ppc-64 and mips bots are still failing after I reverted this change. Since: 1. The mips bot always performs a clean build, 2. The ppc64-bot failed again after a clean build (I asked the ppc-64 maintainers to clean the bot which they did... Thanks Will!), I think it is safe to assume that this change was not the cause of the failures that said builders were seeing. Thus I am recomitting. llvm-svn: 185111
* Revert "[APFloat] Removed APFloat constructor which initialized to either ↵Michael Gottesman2013-06-271-4/+4
| | | | | | | | | | | | zero/NaN but allowed you to arbitrarily set the category of the float." This reverts commit r185095. This is causing a FileCheck failure on the 3dnow intrinsics on at least the mips/ppc bots but not on the x86 bots. Reverting while I figure out what is going on. llvm-svn: 185099
* [APFloat] Removed APFloat constructor which initialized to either zero/NaN ↵Michael Gottesman2013-06-271-4/+4
| | | | | | | | | | | | | | but allowed you to arbitrarily set the category of the float. The category which an APFloat belongs to should be dependent on the actual value that the APFloat has, not be arbitrarily passed in by the user. This will prevent inconsistency bugs where the category and the actual value in APFloat differ. I also fixed up all of the references to this constructor (which were only in LLVM). llvm-svn: 185095
* In InstCombine{AddSub,MulDivRem} convert APFloat.isFiniteNonZero() && ↵Michael Gottesman2013-06-262-5/+5
| | | | | | !APFloat.isDenormal => APFloat.isNormal. llvm-svn: 185037
* [APFloat] Converted all references to APFloat::isNormal => ↵Michael Gottesman2013-06-192-10/+10
| | | | | | | | APFloat::isFiniteNonZero. Turns out all the references were in llvm and not in clang. llvm-svn: 184356
* Simplify code. No functionality change.Jakub Staszak2013-06-061-2/+1
| | | | llvm-svn: 183461
* Re-apply "Use IRBuilder instead of ConstantInt methods." with the fixed issues.Jakub Staszak2013-06-061-68/+55
| | | | llvm-svn: 183439
* Revert "Use IRBuilder instead of ConstantInt methods. It simplifies code a ↵Rafael Espindola2013-06-061-56/+70
| | | | | | | | little bit." This reverts commit 183328. It caused pr16244 and broke the bots. llvm-svn: 183422
* Remove unneeded cast<>.Jakub Staszak2013-06-061-2/+2
| | | | llvm-svn: 183363
* Use IRBuilder instead of ConstantInt methods.Jakub Staszak2013-06-061-27/+17
| | | | llvm-svn: 183360
* Use IRBuilder instead of ConstantInt methods. It simplifies code a little bit.Jakub Staszak2013-06-051-70/+56
| | | | llvm-svn: 183328
* Delete dead safety check.Nick Lewycky2013-06-031-6/+1
| | | | llvm-svn: 183167
* When determining the new index for an insertelement, we may not assume that anNick Lewycky2013-06-011-7/+9
| | | | | | | | | | index greater than the size of the vector is invalid. The shuffle may be shrinking the size of the vector. Fixes a crash! Also drop the maximum recursion depth of the safety check for this optimization to five. llvm-svn: 183080
* Simplify multiplications by vectors whose elements are powers of 2.Rafael Espindola2013-05-311-16/+48
| | | | | | Patch by Andrea Di Biagio. llvm-svn: 183005
* Reapply with r182909 with a fix to the calculation of the new indices forNick Lewycky2013-05-312-2/+257
| | | | | | insertelement instructions. llvm-svn: 182976
* Revert r182909.Evgeniy Stepanov2013-05-302-246/+0
| | | | | | PR/16177 llvm-svn: 182919
* Swizzle vector inputs if it helps us eliminate shuffles.Nick Lewycky2013-05-302-0/+246
| | | | llvm-svn: 182909
* Replace Count{Leading,Trailing}Zeros_{32,64} with count{Leading,Trailing}Zeros.Michael J. Spencer2013-05-241-1/+1
| | | | llvm-svn: 182680
* Run clang-format over the scalarizePHI function.Joey Gouly2013-05-241-12/+8
| | | | llvm-svn: 182640
* scalarizePHI needs to insert the next ExtractElement in the same blockJoey Gouly2013-05-241-2/+4
| | | | | | | | as the BinaryOperator, *not* in the block where the IRBuilder is currently inserting into. Fixes a bug where scalarizePHI would create instructions that would not dominate all uses. llvm-svn: 182639
* This is an update to a previous commit (r181216).Jean-Luc Duprat2013-05-222-29/+43
| | | | | | | | | | | | | | | | | | | | | | | The earlier change list introduced the following inst combines: B * (uitofp i1 C) —> select C, B, 0 A * (1 - uitofp i1 C) —> select C, 0, A select C, 0, B + select C, A, 0 —> select C, A, B Together these 3 changes would simplify : A * (1 - uitofp i1 C) + B * uitofp i1 C down to : select C, B, A In practice we found that the first two substitutions can have a negative effect on performance, because they reduce opportunities to use FMA contractions; between the two options FMAs are often the better choice. This change list amends the previous one to enable just these inst combines: select C, B, 0 + select C, 0, A —> select C, B, A A * (1 - uitofp i1 C) + B * uitofp i1 C —> select C, B, A llvm-svn: 182499
* Add missing -*- C++ -*- to headersMatt Arsenault2013-05-171-1/+1
| | | | llvm-svn: 182164
* Fix two typoSylvestre Ledru2013-05-141-1/+1
| | | | llvm-svn: 181848
* InstCombine: Flip the order of two urem transformsDavid Majnemer2013-05-121-6/+6
| | | | | | | | | | | | | | There are two transforms in visitUrem that conflict with each other. *) One, if a divisor is a power of two, subtracts one from the divisor and turns it into a bitwise-and. *) The other unwraps both operands if they are surrounded by zext instructions. Flipping the order allows the subtraction to go beneath the sign extension. llvm-svn: 181668
* InstCombine: Turn urem to bitwise-and more oftenDavid Majnemer2013-05-111-20/+2
| | | | | | | Use isKnownToBeAPowerOfTwo in visitUrem so that we may more aggressively fold away urem instructions. llvm-svn: 181661
* InstCombine: Don't claim to be able to evaluate any shl in a zexted type.Benjamin Kramer2013-05-101-1/+11
| | | | | | | | | | The shift amount may be larger than the type leading to undefined behavior. Limit the transform to constant shift amounts. While there update the bits to clear in the result which may enable additional optimizations. PR15959. llvm-svn: 181604
* InstCombine: Verify the type before transforming uitofp into select.Benjamin Kramer2013-05-101-22/+23
| | | | | | PR15952. llvm-svn: 181586
* InstCombine: Don't just copy known bits from the first operand of an srem.Benjamin Kramer2013-05-091-1/+1
| | | | | | | That's obviously wrong. Conservatively restrict it to the sign bit, which matches the original intention of this analysis. Fixes PR15940. llvm-svn: 181518
* InstCombine: (X ^ signbit) + C -> X + (signbit ^ C)David Majnemer2013-05-061-0/+5
| | | | llvm-svn: 181249
* Provide InstCombines for the following 3 cases:Jean-Luc Duprat2013-05-062-0/+53
| | | | | | | | | | | | | A * (1 - (uitofp i1 C)) -> select C, 0, A B * (uitofp i1 C) -> select C, B, 0 select C, 0, A + select C, B, 0 -> select C, B, A These come up in code that has been hand-optimized from a select to a linear blend, on platforms where that may have mattered. We want to undo such changes with the following transform: A*(1 - uitofp i1 C) + B*(uitofp i1 C) -> select C, A, B llvm-svn: 181216
* Revert r164763 because it introduces new shuffles.Nadav Rotem2013-05-061-19/+1
| | | | | | Thanks Nick Lewycky for pointing this out. llvm-svn: 181177
* Add ArrayRef constructor from None, and do the cleanups that this ↵Dmitri Gribenko2013-05-051-1/+1
| | | | | | | | constructor enables Patch by Robert Wilhelm. llvm-svn: 181138
* Tabs to spaces. No functionality change.Nick Lewycky2013-05-041-3/+3
| | | | llvm-svn: 181082
* This patch breaks up Wrap.h so that it does not have to include all of Filip Pizlo2013-05-011-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | the things, and renames it to CBindingWrapping.h. I also moved CBindingWrapping.h into Support/. This new file just contains the macros for defining different wrap/unwrap methods. The calls to those macros, as well as any custom wrap/unwrap definitions (like for array of Values for example), are put into corresponding C++ headers. Doing this required some #include surgery, since some .cpp files relied on the fact that including Wrap.h implicitly caused the inclusion of a bunch of other things. This also now means that the C++ headers will include their corresponding C API headers; for example Value.h must include llvm-c/Core.h. I think this is harmless, since the C API headers contain just external function declarations and some C types, so I don't believe there should be any nasty dependency issues here. llvm-svn: 180881
* Revert "InstCombine: Fold more shuffles of shuffles."Jim Grosbach2013-05-011-12/+5
| | | | | | | | | This reverts commit r180802 There's ongoing discussion about whether this is the right place to make this transformation. Reverting for now while we figure it out. llvm-svn: 180834
* InstCombine: Fold more shuffles of shuffles.Jim Grosbach2013-04-301-5/+12
| | | | | | | | | | | Always fold a shuffle-of-shuffle into a single shuffle when there's only one input vector in the first place. Continue to be more conservative when there's multiple inputs. rdar://13402653 PR15866 llvm-svn: 180802
* Fix a bug in foldSelectICmpAndOr.David Majnemer2013-04-301-1/+2
| | | | | | | Differences in bitwidth between X and Y could exist even if C1 and C2 have the same Log2 representation. llvm-svn: 180779
* Fix "Combine bit test + conditional or into simple math"David Majnemer2013-04-301-0/+64
| | | | | | | | | This fixes the optimization introduced in r179748 and reverted in r179750. While the optimization was sound, it did not properly respect differences in bit-width. llvm-svn: 180777
* Move C++ code out of the C headers and into either C++ headersEric Christopher2013-04-221-0/+1
| | | | | | | or the C++ files themselves. This enables people to use just a C compiler to interoperate with LLVM. llvm-svn: 180063
* Changed back (relative to commit 179786) the operations executed when ↵Anat Shemer2013-04-221-3/+3
| | | | | | extract(cast) is transformed to cast(extract). It uses the Builder class as before. In addition the result node is added to the Worklist, so all the previous extract users will become the new scalar cast users. llvm-svn: 180045
OpenPOWER on IntegriCloud